Schapiro ’19: Use the right stats, not the common ones

For the most part, baseball writers watched and described baseball the same way from about 1860 to 2001, give or take a few years either way. This isn’t entirely a bad thing: without all the stories of early baseball that these writers recorded, which may well otherwise have been forgotten, baseball wouldn’t have half the charm and tradition that it does today.

But the fact that writers went more than a century without adapting their coverage and analysis to developments in scouting and statistics has also left modern fans with a burden. Coverage of baseball has worked the same way for so long that fans are only now confronting the ultimate is/ought fallacy — what fans have been told for centuries isn’t actually how the game works.

There’s the walk, for example. Early baseball writers had a flawed understanding of strategy and statistics, which meant that for more than a hundred years, writers sorted players based on batting average. On-base percentage, which includes walks as well as hits, is the vastly more important statistic, but we’d been using batting average for so long that people thought it couldn’t just be wrong.

The same is true of the runs batted in, although as a statistic, RBI seems to be losing much of its undeserved clout in recent years. It’s easy to see what’s wrong with RBI: The best hitter in the world could play for a team of incompetents and might not drive in as many runs as a mediocre hitter on a team full of superstars who are consistently on base. But RBI caught on early because it makes intuitive sense. It’s hard to argue with getting a hit that drives in a run, even though it turns out that statistically speaking, RBIs are neither predictive nor comparative. Driving in runs now doesn’t mean a player is likely to drive in runs later, and it certainly doesn’t mean that a player is a good hitter.

On the defensive side, there’s the error, another statistic which must have made sense at first, and as such went unchallenged for more than a century. But the problem with the error is that it’s completely subjective and far too limited in scope. To be assigned an error in the first place, a player has to be quick enough to get in position to field a ball. Meanwhile, a sub-par defender might not rack up errors for the simple reason that they weren’t in position to make difficult plays. If you sprint for a ball and it bounces out of your glove, that’s an error — if you jog and pick it up on two bounces, it’s a clean single.

And then, of course, there’s the eternal one-two punch of misleading pitching statistics, the win and the save. As far as saves go, I’m actually less skeptical. Pitching the ninth inning is no more valuable, statistically speaking, than pitching the seventh or the fourth; depending on the situation in any inning (which can be measured by a statistic called leverage index), any inning might be more important than the ninth. But there’s something about being able to pitch the ninth inning and nail down a win … there’s some value to being able to do that, even though the data may not show it. And that value is what the save measures, albeit imperfectly. So the save, until we find a better way to measure that intangible closer’s mentality, still has some value, although not as much as most people think.

But the win — well, that’s another story. The win has been losing its stature as a defining pitching statistic since 2010, when the Baseball Writers’ Association of America, in recognition of what it could tell was a stellar season, awarded Félix Hernández the Cy Young award despite a mediocre 13-12 record. And this year, the win looks ready to die completely. As evidence, I give you New York, where Mets pitcher Jacob deGrom is having one of the best pitching seasons in recent memory. DeGrom’s ERA is 1.71; his ERA+ is 216; on average, he is allowing less than a baserunner per inning and striking out 11 batters per nine.

But the Mets’ offense is a farce, a sick joke, so deGrom doesn’t look like a winning pitcher: his record is 8-9. In one of the greatest pitching seasons in a long time, deGrom, at least by win-loss record, looks like a sub-average pitcher.

It’s not hard to see how much the Mets’ paltry offense has hurt deGrom. In his eight wins this season, deGrom has pitched to a 0.97 ERA; in his other 21 outings, his ERA is 2.20. Put another way, take away deGrom’s eight winning starts, and his record would be 0-9 — and his 2.20 ERA would still be the best in the National League. The win is dead. The Mets’ offense killed it.

Why does all of this matter? Because awards season is coming up. And as baseball fans everywhere debate their picks for Most Valuable Player, Cy Young and Rookie of the Year, we should remember what’s actually important. It’s important not to form opinions based on numbers if we don’t know why the numbers are meaningful, or if we only think the numbers are important because they’re what we’re accustomed to using. Baseball is evolving, finally, and our understanding of the game is evolving too. So before making a point about a player, it’s essential to understand the numbers behind the point, and to know whether they actually mean anything.

James Schapiro ‘19 can be reached at Please send responses to this opinion to and op-eds to

Read more here:
Copyright 2018